I like the way you framed this. What you’re getting at is the difference between GTO vs exploitive play. Of course it’s best to know exactly what your opponents are playing, but that is rarely the case. Poker is a game of incomplete information, and game theory is a way of determining the best play when your opponents’ strategies are not known to you (or when they are not static, as with poker – someone who was formerly tight could suddenly start playing more loosely, either deliberately to exploit that image or just because he is bored/frustrated). It’s good to have an idea of which plays are profitable under conditions of uncertainty and which require reads to justify. I don’t think the default approach that, at least a few years ago, was popular among MTT specialists, of only playing very strong hands in the early stages of a tournament, is optimal. It may, however, be good if you are comfortable making the default assumption that your opponents will play too loosely, especially pre-flop, and will not hand read well or make disciplined folds. But against better players, you will face a lot of tough decisions, rarely put them to tough decisions, and rarely win big pots with your strong hands.